We study environmental effects on human reproduction, together with the basic biology that underlies such effects. We recently established that human conception occurs with intercourse during a six-day interval of the menstrual cycle. This interval ends on the day of ovulation, suggesting that human sperm can survive up to five days in the female tract, while the ovum is fertilizable for less than a day. Based on these data we have developed statistical methods to estimate the chance of conception on each day of the cycle, taking into account possible effects of reproductive toxins. Contrary to popular belief, we found no evidence that couples can influence the sex of their baby by timing their intercourse around ovulation. However, there was intriguing evidence that male babies are more often conceived in cycles with short follicular phases. In another application of previously developed methods, we observed a seasonal effect on pregnancy loss, raising the possibility of environmental effects on such loss. We have continued to explore the etiology of reproductive failure. Surprisingly, we could find no association between the poor conditions suffered in Norway under Nazi occupation and perinatal mortality. In a further exploration of reasons for lower infant mortality in Norway compared to the US, we identified an excess of preterm delivery as am important component of US infant deaths. A separate study found that long periods of standing or walking during pregnancy may increase the risk of preterm delivery. Finally, in a major study of adults who were prenatally exposed to DES, we were unable to confirm a variety of health effects seen in animal studies and reported anecdotally in humans. DES sons were found to have the same fertility as unexposed sons. DES daughters showed no evidence of premature menopause. Finally, we found no evidence of higher rates of infection or autimmune disease among DES offspring. This suggests that humans may be less vulnerable to adverse effects of prenatal estrogens than has been hypothesized.